Measuring disease: incidence and prevalence
Introduced| English | Meaning |
|---|---|
| incidence/ˈɪnsɪdəns/ | incidence |
| prevalence/ˈprevələns/ | prevalence |
| case fatality | case fatality |
| age standardisation | age standardisation |
Two towns, one disease
- Town A reports twice the death rate of Town B from the same disease. Is Town A doing worse? Not if Town A simply has more old people.
- Before comparing anything, epidemiology counts disease carefully and matches each number to the right denominator.
Incidence: new cases
- Incidence 发病率 = new cases ÷ people at risk ÷ time. It measures how fast disease appears, for example 2% per year.
- Only people who could get the disease count in the denominator; people who already have it are not at risk.
(Synthetic) 1,500 people at risk; 30 new cases in one year. What is the incidence, in % per year?
Difficulty 2/5 · 140 coins + a small bonus on first completion today
30 ÷ 1,500 = 0.02, so 2% per year.
Prevalence: existing cases
- Prevalence 患病率 = existing cases ÷ population at one moment. It is a proportion with no time unit.
- Prevalence ≈ incidence × average duration. Long illnesses build up; short ones clear between surveys.
Prevalence is a rate with a time unit.
Difficulty 1/5 · 100 coins + a small bonus on first completion today
It is a proportion at one moment; incidence is the rate.
(Synthetic) Incidence is 0.5% per year and the illness lasts 4 years on average. Estimate the prevalence in %.
Difficulty 2/5 · 140 coins + a small bonus on first completion today
P ≈ I × D = 0.5% × 4 = 2%.
Three ways to count deaths
- Case fatality 病死率 = deaths ÷ cases: how deadly the disease is for those who have it.
- Cause-specific mortality = deaths ÷ whole population per year. Proportional mortality = this disease's share of all deaths.
Match each measure to its denominator.
Difficulty 2/5 · 140 coins + a small bonus on first completion today
The denominator decides the question being answered.
(Synthetic) 48 people catch a disease and 6 die. What is the case fatality in %?
Difficulty 2/5 · 140 coins + a small bonus on first completion today
6 ÷ 48 = 0.125, so 12.5%.
Age standardisation
- Crude rates mix disease risk with age structure. A town with more old people can show a higher crude rate even with the same risk at every age.
- Age standardisation 年龄标化 applies each town's age-specific rates to one standard population, so the comparison uses the same age mix.
Put the steps of direct age standardisation in order.
Difficulty 3/5 · 200 coins + a small bonus on first completion today
Both towns now share the same age mix.
Worked example: a synthetic town
- (Synthetic) 2,000 residents; 40 new cases in a year: incidence 40/2,000 = 2% per year. On survey day 60 have the disease: prevalence 3%.
- Check: 2% per year × 1.5 years average duration = 3%. Of the 40 cases, 5 died: case fatality 5/40 = 12.5%.
Worked example: same rates, different crude
- (Synthetic) Rates are 1 per 1,000 for the young and 10 per 1,000 for the old in both towns. Town A: 1,000 young and 1,000 old: (1 + 10) ÷ 2,000 = 5.5 per 1,000.
- Town B: 1,800 young and 200 old: (1.8 + 2) ÷ 2,000 = 1.9 per 1,000. The risk is identical; only the age mix differs.
Two towns have the same age-specific rates, but one has far more old people. What do the crude rates show?
Difficulty 1/5 · 100 coins + a small bonus on first completion today
More people in the high-risk age group raises the crude rate.
Three common slips
- Calling prevalence a rate. It is a proportion at one moment.
- Comparing crude rates across different age structures. Standardise first.
- Reading case fatality as mortality. Its denominator is cases, not the population.
Recap
- Incidence = new ÷ at risk ÷ time; prevalence = existing ÷ population; P ≈ I × D.
- Case fatality (÷ cases), cause-specific mortality (÷ population), proportional mortality (÷ all deaths).
- Standardise for age before comparing populations.